基于无人机的大场景序列图像自动采集和三维建模

发布时间：2018-02-05 01:10

本文关键词： 无人机序列图像三维建模自动采集 SFM　出处：《西北大学学报(自然科学版)》2017年01期 　论文类型：期刊论文

【摘要】：提出并实现了一种基于无人机的大场景序列图像自动采集和三维重建方法,实现了大范围场景全自动图像数据采集和三维建模。首先设计实现无人机自动控制地面站和基于Cortex-A17的无人机控制模块,其次在地面站人工点选设计航线和拍摄位置,并自动控制飞行和拍摄过程,实时将序列图像回传地面站,最后地面站应用SFM方法实时自动重建大场景三维模型。实验结果表明,相对于人工控制方法,文中方法能够实现无人机飞行和拍摄过程自动控制,同时提高拍摄稳定性,降低基于无人机的大场景三维建模难度,提高建模效果。
[Abstract]:An automatic image acquisition and 3D reconstruction method for large scene sequences based on UAV is proposed and implemented. Automatic scene image data acquisition and 3D modeling are realized. Firstly, the UAV automatic control ground station and UAV control module based on Cortex-A17 are designed and implemented. Secondly, the design route and shooting location are manually selected at the ground station, and the flight and shooting process are automatically controlled, and the sequence images are transmitted back to the ground station in real time. Finally, the SFM method is used to reconstruct the 3D model of large scale scene automatically. The experimental results show that compared with the manual control method, the proposed method can realize the automatic control of UAV flight and shooting process. At the same time, it improves the shooting stability, reduces the difficulty of 3D modeling of large scene based on UAV, and improves the modeling effect.
【作者单位】：西北大学信息科学与技术学院;西北大学文化遗产数字化国家地方联合工程研究中心;
【基金】：国家自然科学基金资助项目(61373117) 陕西省教育厅专项科研计划基金资助项目(16JK1775)
【分类号】：TP391.41
【正文快照】： 随着各个行业信息化程度的提高,使用三维模型准确记录现实世界大规模场景表面几何和纹理形态的需求日趋迫切。传统的人工测量拍照并数字化建模的方法,由于费时费力并且准确性,已经很少被应用在大场景三维建模中;20世纪80年代以来,应用基于激光或相位技术的非接触式扫描设备对

【参考文献】